Meta-Analysis of Antibiotics and the Risk of Community-Associated Clostridium difficile Infection
ABSTRACT
The rising incidence of Clostridium difficile infection (CDI) could be reduced by lowering exposure to high-risk antibiotics. The objective of this study was to determine the association between antibiotic class and the risk of CDI in the community setting. The EMBASE and PubMed databases were queried without restriction to time period or language. Comparative observational studies and randomized controlled trials (RCTs) considering the impact of exposure to antibiotics on CDI risk among nonhospitalized populations were considered. We estimated pooled odds ratios (OR) for antibiotic classes using random-effect meta-analysis. Our search criteria identified 465 articles, of which 7 met inclusion criteria; all were observational studies. Five studies considered antibiotic risk relative to no antibiotic exposure: clindamycin (OR = 16.80; 95% confidence interval [95% CI], 7.48 to 37.76), fluoroquinolones (OR = 5.50; 95% CI, 4.26 to 7.11), and cephalosporins, monobactams, and carbapenems (CMCs) (OR = 5.68; 95% CI, 2.12 to 15.23) had the largest effects, while macrolides (OR = 2.65; 95% CI, 1.92 to 3.64), sulfonamides and trimethoprim (OR = 1.81; 95% CI, 1.34 to 2.43), and penicillins (OR = 2.71; 95% CI, 1.75 to 4.21) had lower associations with CDI. We noted no effect of tetracyclines on CDI risk (OR = 0.92; 95% CI, 0.61 to 1.40). In the community setting, there is substantial variation in the risk of CDI associated with different antimicrobial classes. Avoidance of high-risk antibiotics (such as clindamycin, CMCs, and fluoroquinolones) in favor of lower-risk antibiotics (such as penicillins, macrolides, and tetracyclines) may help reduce the incidence of CDI.
INTRODUCTION
Clostridium difficile, a toxin-producing bacterium that causes diarrhea, is the largest single cause of morbidity and mortality among hospital-acquired infections (1). In hospitals, C. difficile infection (CDI) is generally acquired when patients with predisposing factors such as advanced age and antibiotic use are exposed to C. difficile spores emanating from other hospitalized infected patients (2). With the emergence of increasingly virulent C. difficile strains have come reports of CDIs in patients previously considered to be at low risk of this infection, including those living in the community (3–5). Spore exposure may occur outside inpatient settings, since river water, soil, and foods can be contaminated (6, 7), outpatient exposures to the health care system are common, and transmission may occur within households (8). A recent study noted that the population-based incidence of community-acquired CDI (11.2 cases per 100,000 person-years) was on par with hospital-acquired CDI (12.1 cases per 100,000 person-years) (9).
One published meta-analysis and one systematic review have considered the impact of antibiotic exposure on CDI (10, 11) risk among hospital inpatients. The meta-analytic study noted that tetracyclines and penicillins were associated with the lowest risk, while fluoroquinolones, clindamycin, and expanded-spectrum cephalosporins were associated with the highest risk of CDI acquisition, despite considerable confidence interval overlap (10). The systematic review established that the strongest evidence of risk existed for penicillins and clindamycin and that effect estimates for other antibiotic classes were liable to bias (11).
In addition to yielding accurate adjusted effect estimates, a systematic review of the association between exposure to antibiotics and community-associated CDI is necessary, since the risk profile is different among nonhospitalized populations (younger, less frequent exposure to patients with symptomatic CDI, and different profile of underlying infections and antibiotic treatments). We conducted a systematic review of the association between antibiotic type and the risk of CDI in nonhospitalized populations. Our objective was to quantify the relative risks of specific antibiotics in order to better understand the risks of prescribing various antibiotics in the community setting.
MATERIALS AND METHODS
Search criteria.
A literature search was conducted in March 2012 using the EMBASE and PubMed databases and included all articles without restriction to language or time period. The Reference sections of the articles were browsed, and content experts were approached to identify further relevant articles. Within each database, our search strategy was to use both key words and mapped subject headings as terms describing the exposure (i.e., antibiotic, antibacterial, antimicrobial, aminoglyosides, beta-lactams, cephalosporins, clindamycin, fluoroquinolones, macrolides, metronidazole, sulfonamides, and tetracyclines), outcome (C. difficile infection), and the detection of a community-acquired infection (community-acquired, community-associated, outpatient, ambulatory care, registry, and general practice). Exposure, outcome, and population terms were then combined using the Boolean “and” operator (12).
We included population-based studies of people with little to no health care system contact prior to the onset of disease (13); studies restricted to hospital-acquired or health care-associated disease (e.g., studies restricted to HIV and cancer outpatients) were excluded. The outcome of interest was incident CDI (collected at the individual level). We were interested in exposure to specific antibiotic classes. We excluded studies considering risk factors for severe or relapsing CDI among individuals already diagnosed with C. difficile, time series analyses, and those not examining specific antibiotics or antibiotic classes.
Screening and data abstraction.
One author (K. A. Brown) screened article titles and abstracts of the initial database search to identify those appropriate for full text review. The full text of identified articles was read; those articles eliminated in the initial screen of titles and abstracts were distinguished from articles eliminated in the full-text screen. Data on the numbers of cases and controls, unadjusted and adjusted effect sizes, and 95% confidence intervals corresponding to each antibiotic exposure group reported were abstracted and entered into a spreadsheet. When insufficient information was available to obtain the appropriate effect size standard errors, study authors were contacted by e-mail.
Quality assessment.
Study quality was assessed using a six-criterion checklist (14–16) with certain elements of the checklist aimed at addressing specific concerns raised in previous systematic reviews on the topic (10, 11). The six study quality assessment questions follow. (Question 1 [Q1]) Did the study have a clearly stated study aim? (Q2) Was the study population clearly defined (i.e., was an appropriate method used to ensure that inpatients were excluded)? (Q3) Were the case counts of the antibiotic exposure groups reported, and if so, were they sufficiently large to statistically compare effect sizes? A study with 10 or more cases in each antibiotic exposure group was given a score of 2, <10 in some groups was given a score of 1, and a study not reporting case counts was given a score of 0. (Q4) Were the criteria for diagnosis of C. difficile clearly defined (e.g., statement of ICD [international classification of diseases] code for registry studies; clear description of microbiologic methods used for clinical studies)? (Q5) Was there an attempt to use statistical analyses to adjust for confounding or to standardize disease rates by age, duration of antibiotic exposure, and exposure to combinations of antibiotics? (Q6) Was there a discussion of study limitations? Each quality criterion identified except question 3 was scored as 0 (no) or 1 (yes); question 3 was graded on a scale from 0 to 2. Two authors (K. A. Brown and N. Khanafer) independently graded study quality; the results were compared, and disagreements were resolved through discussion. By summing up the points, the studies were classified as high quality (6 or 7 points), moderate quality (4 or 5 points), or poor quality (<4 points).
Variables.
Antibiotics were classified into one of the following 7 groups from the articles included in this meta-analysis study: (i) tetracyclines; (ii) sulfonamides and trimethoprim; (iii) penicillins; (iv) macrolides; (v) cephalosporins, monobactams, and carbapenems (CMCs); (vi) fluoroquinolones; and (vii) clindamycin. The dependent variable of interest was the adjusted study-specific log odds ratio of a given antibiotic class relative to no antibiotic exposure; this variable and the standard error for each reported effect were usually transcribed directly from the publication Results section. In the study of Dial et al. (17), the effect sizes of levofloxacin, gatifloxacin, and moxifloxacin antibiotics were combined by taking the weighted average of the log odds ratios, with inverse variance weights; in the study of Kuntz et al. (9), the effect sizes for the CMCs were combined in a similar manner.
Statistical analysis.
A pooled random-effects analysis was used to consider the impact of any antibiotic exposure irrespective of antibiotic class relative to no exposure, using the DerSimonian-Laird approach (18). A stratified analysis was then used to consider the risk associated with each antibiotic type compared to no antibiotic exposure. In a secondary analysis restricted to the 2 studies excluding antibiotic unexposed persons (19, 20), odds ratios were calculated from the number of CDI-positive cases and the total person-time within antibiotic exposure classes with 0.5 added to each cell (in order to estimate effects in the presence of zero cells) (21); for the odds ratio, penicillin-exposed persons were considered the referent category. Due to the low incidence of community-acquired CDI in contemporary populations (9), the odds ratios reported in this study may be interpreted as risk ratios (22). For the studies in the secondary analysis, a similar stratified meta-analysis measured the risk associated with each antibiotic type compared to penicillin exposure.
We assessed the heterogeneity of study results by use of the τ2, Q (18), and I2 statistics (23). Possible sources of heterogeneity were explored in sensitivity analyses in which certain subgroups were excluded, as well as through creation of meta-regression models (24). Analyses were conducted in R using the meta and metafor packages (25). Data and R code have been deposited in the Dryad repository (https://doi.org/10.5061/dryad.g7b05).
RESULTS
Out of a total of 465 articles identified, only 7 fulfilled the eligibility criteria (Fig. 1). Three studies employed nested case control designs, 3 used nonnested case control designs, and one was a cohort study (Table 1). The studies monitored subjects from 1988 to 2007 and varied in size from as few as 40 to over 1,200 cases of CDI.
Fig 1
Table 1
| Characteristic | Delaney et al. (34) | Dial et al. (17) | Hirschhorn et al. (19) | Kuntz et al. (9) | Levy et al. (20) | Naggie et al. (36) | Wilcox et al. (27) |
|---|---|---|---|---|---|---|---|
| Publication yr | 2007 | 2008 | 1994 | 2011 | 2000 | 2011 | 2008 |
| Study design | Case control | Nested case control | Retrospective cohort | Nested case control | Nested case control | Case control | Case control |
| Data source(s) | UK General Practice Research Database (GPRD) and British National Formulary | Hospital discharge summary and medical insurance records of Quebec, Canada | Harvard Community Health Plan | University of Iowa Wellmark Data Repository (Iowa and South Dakota, USA) | United Health Group claims database | Duke University Medical Center, Durham Regional Hospital, Durham, Salisbury, and Asheville VAMCsa | UK laboratory test results and telephone questionnaire |
| Study periodb | Jan. 1993 to Dec. 2004 | Jan. 1996 to Dec. 2004 | Apr. 1988 to Nov. 1990 | Jan. 2004 to Dec. 2007 | Jan. 1992 to Dec. 1994 | Oct. 2006 to Nov. 2007 | Jan. to Dec. 1999 |
| Study population | Persons registered in GPRD not hospitalized in the last yr | Persons ≥ 65 yrs old, with at least one previous hospitalization, without hospital discharge in the previous 90 days | Enrollees not hospitalized in the last 42 days and with ≥1 antibiotic prescription in the previous 2 to 42 days | Enrollees with >12 mo of health coverage without discharge in the previous 12 wks | Enrollees with a C. difficile lab test and with exactly 1 antibiotic exposure in the previous 2 to 42 days | Outpatients ≥ 18 yrs old without history of diarrhea and without discharge in the previous 12 wks | Outpatients with a fecal sample sent for C. difficile toxin test |
| Case definition | Presence of an initial positive C. difficile toxin assay result and/or a clinical diagnosis recorded by their general practitioner (GP) | Primary hospital diagnosis of CDI (ICD coding) | Positive assay result for C. difficile toxin with documented diarrhea or colitis with onset within 48 h of admission | Primary or secondary CDI diagnosis on insurance claim. For hospital patients, it must be recorded at admission. | Documented diarrhea and a positive C. difficile laboratory test result in the medical record | Patients with diarrhea (onset in community or ≤72 h of hospital arrival) and a positive stool toxin assay result | Patients presenting to their GP with symptoms of diarrhea with positive stool toxin assay result |
| Matching | Age, clinic site, and index datec | Index datec and date of 1st hospitalization | None | Index datec | None | Index moc and clinic site | Age, sex, and index periodc (3 mo) |
| Antibiotic exposure | National claims database; 90-day window | Provincial claims database; 45-day window | Health maintenance organization (HMO) pharmacy records; 2- to 42-day window | Outpatient drug claims; 180-day window | HMO pharmacy records; 2- to 42-day window | Electronic medical records; 90-day window | Postal questionnaire; 4-wk window |
| Adjustment method | Conditional logistic regression | Conditional logistic regression | Stratification | Conditional logistic regression | Stratification | Logistic regression | Matching |
| No. of cases (n)/no. of controls (N) | 1,233/12,330 | 836/8,360 | 51/175,275d | 304/3,040 | 48/358,389d | 66/114 | 40/112 |
a
VAMC, Veterans Affairs Medical Center.
b
Jan., January; Dec., December; Apr., April; Nov., November; Oct., October.
c
The index date (or month) was defined as the date of diagnosis, symptom onset, or admission for cases and the date of health care facility visit for controls.
d
N = person-days.
A total of 5 studies included controls without antibiotic exposure and could be included in the primary meta-analyses; the other 2 studies were included in a secondary meta-analysis limited to studies without antibiotic-free controls.
Of the studies in the primary analysis, two evaluated all 7 candidate antibiotic classes, two covered between 5 and 6 of the 7 classes, and one reported on only 4 of 7 classes. The two studies in the secondary analysis each reported exposures for all 7 antibiotic classes, but no patients exposed to clindamycin in either of these additional studies acquired CDI infection, so the odds ratios were not calculated for this agent.
Among the studies included in the primary meta-analyses, study quality (Table 2) was scored high for two studies, moderate for two studies, and low for one study. Among studies included in the secondary meta-analyses, one was scored as high quality, and the other was scored as low quality. Note that for three studies, the case definition did not properly exclude potentially hospital-acquired cases (17, 26, 27).
Table 2
| Study (author, reference, and publication yr) | Quality scorea | ||||||
|---|---|---|---|---|---|---|---|
| Total (0–7) | Q1 | Q2 | Q3b | Q4 | Q5 | Q6 | |
| Delaney et al. (34) (2007) | 6 | 1 | 1 | 2 | 0 | 1 | 1 |
| Dial et al. (17) (2008) | 4 | 1 | 1 | 0 | 0 | 1 | 1 |
| Hirschhorn et al. (19) (1994) | 6 | 1 | 1 | 1 | 1 | 1 | 1 |
| Kuntz et al. (9) (2011) | 7 | 1 | 1 | 2 | 1 | 1 | 1 |
| Levy et al. (20) (2000) | 3 | 1 | 0 | 1 | 0 | 0 | 1 |
| Naggie et al. (36) (2011) | 4 | 1 | 0 | 0 | 1 | 1 | 1 |
| Wilcox et al. (27) (2008) | 3 | 1 | 0 | 0 | 1 | 0 | 1 |
a
The six study quality assessment questions and scoring method are given in “Quality assessment” in Materials and Methods.
b
Scored on a scale from 0 to 2.
Pooled effects.
The pooled impact of any antibiotic exposure (irrespective of antibiotic class) across all 29 antibiotic effects (Fig. 2) was to increase the risk of CDI by a multiple of 3 (OR = 3.55; 95% CI, 2.56 to 4.94). In this analysis, which ignored antibiotic class, effect heterogeneity was extremely high (I2 = 90.6%; P < 0.001).
Fig 2
Antibiotic types.
In the analyses stratified by antibiotic class, 6 of 7 antibiotic classes were associated with increased risk of CDI (Fig. 2). Specifically, clindamycin (OR = 16.80; 95% CI, 7.48 to 37.76), fluoroquinolones (OR = 5.50; 95% CI, 4.26 to 7.11), and CMCs (OR = 5.68; 95% CI, 2.12 to 15.23) were found to increase CDI risk the most, while macrolides (OR = 2.65; 95% CI, 1.92 to 3.64), sulfonamides and trimethoprim (OR = 1.81; 95% CI, 1.34 to 2.43), and penicillins (OR = 2.71; 95% CI, 1.75 to 4.21) were found to have a lesser, but nevertheless statistically significant impact. There was no evidence of the impact of tetracyclines on CDI risk (OR = 0.92; 95% CI, 0.61 to 1.40).
Between-study heterogeneity was largest in the CMC (I2 = 93.8%; P < 0.001), penicillin (I2 = 76.8%; P = 0.002), and clindamycin (I2 = 66.7%; P = 0.05) drug classes. Conversely, heterogeneity was lowest for tetracyclines (I2 = 0.0%; P = 0.98), sulfonamides and trimethoprim (I2 = 0.0%; P = 0.56), and fluoroquinolones (I2 = 10.9%; P = 0.34). Relative to the pooled meta-analysis (τ2 = 0.62), the stratified model reduced heterogeneity by 55% (τ2 = 0.27; P < 0.001).
Meta-regression.
Meta-regression was used in order to investigate the factors influencing residual heterogeneity from the primary analysis. The five studies were associated with systematic differences in drug effects (P = 0.001 by the χ24 test); in particular, the pooled odds ratios in one study (17) were twice those of the remaining studies (OR = 1.93; 95% CI, 1.30 to 2.87). The addition of study level effects to the meta-regression model reduced heterogeneity by 63% (τ2 = 0.10; P < 0.001). As a sensitivity analysis, we excluded the one study reporting larger effect sizes; the pooled odds ratio in the remaining 4 studies was 2.86 (95% CI, 2.86 to 3.81) and the between-study effects were no longer significant (P = 0.12 by the χ23 test).
For the subset of high-quality studies (n = 2), the heterogeneity of between-study effects was below detectable limits (P = 0.96 by the χ21 test). These two studies reported effect sizes for all antibiotic classes that were smaller than those of medium- and low-quality studies (OR = 2.50; 95% CI 1.80 to 3.47).
Publication bias.
We tested for funnel plot asymmetry using the stratified model and found no evidence of an association between effect estimate precision and residual effect sizes (z = 0.53; P = 0.59).
Antibiotic-associated CDI risk (AACR) index.
In a post hoc exploratory analysis, a simple 4 point index summarizing the meta-analysis results was developed; the index was equal to 1 for tetracyclines, 2 for sulfonamides and trimethoprim, macrolides, and penicillins, 3 for CMCs and fluoroquinolones, and 4 for clindamycin. Each one-point increase in the index was associated with a 2.41-fold increase (95% CI, 2.14 to 2.74) in the odds of acquiring CDI. Mean antibiotic class effect did not deviate significantly from the linear trend (P = 0.30 for sulfanomides to P = 0.85 for tetracyclines). The model fit is presented graphically in Fig. 3.
Fig 3
Secondary analysis.
With this analysis, similar findings were noted; namely, tetracyclines (OR = 0.60; 95% CI, 0.14 to 2.61) were not associated with an increased risk of C. difficile; sulfonamides and trimethoprim (OR = 0.85; 95% CI, 0.29 to 2.52), and macrolides (OR = 0.60; 95% CI, 0.20 to 1.76) tended to have the smallest effect sizes with the least heterogeneity, while CMCs (OR = 4.12; 95% CI, 2.28 to 7.44) and fluoroquinolones (OR = 4.31; 95% CI, 1.46 to 12.70) had larger and more variable effect sizes. In both studies, clindamycin exposure was rare (<0.5% of total antibiotic exposures in each); neither reported any cases associated with clindamycin.
DISCUSSION
The emergence of C. difficile as an infection in individuals without prior hospitalization, and presumed community acquisition, represents a concerning development in the ongoing emergence of this pathogen. As any prescription of an antimicrobial agent to a patient in an outpatient setting requires a careful weighing of risks and benefits, we performed a systematic review to quantify the risks associated with individual antibiotic classes and to identify areas of heterogeneity in such risk. Overall use of antibiotic agents is associated with a 3-fold increased risk of community-acquired CDI, but we also detected substantial variation in risk associated with different antimicrobial classes, with fluoroquinolones, CMCs, and clindamycin associated with the greatest enhancement of risk.
This study largely corroborates the associations found for hospital-associated CDI risk (10, 11, 28). In keeping with many historic studies of CDI risk and outbreaks of the disease, clindamycin was found to have the strongest association with risk. One must note however, that clindamycin has not been associated with the greatest risk enhancement in every study (28); variability in effects may be due to true biological heterogeneity of effect (e.g., variable strain susceptibility to clindamycin [29], timing of inoculation relative to the end of antibiotic exposure), or it could be an artifact of the different methods used for outcome ascertainment (see below).
Our study found large effects for fluoroquinolones and CMCs. This could be expected given the broad spectrum of activity of these agents against intestinal microbes and the low susceptibility of Clostridium difficile to these classes of antibiotics (30). The risk associated with CMCs was highly variable across studies, in contrast to fluoroquinolones, which appeared to have more-consistent effects. This heterogeneity may be due to the greater activity of newer cephalosporins against anaerobic bacteria and Gram-negative bacilli (10). In contrast, one study limited to patients with fluoroquinolone exposures (31) found no differences in effect between levofloxacin, gatifloxacin, and moxifloxacin, notwithstanding the enhanced anti-anaerobic spectrum of the latter agents.
Our findings reaffirmed the finding of moderate effects for penicillins, macrolides, and sulfonamides and trimethoprim from a recent hospital-based study (28); this is in contrast to other hospital-based studies that have noted large effects for penicillins (11). The relatively low MICs for penicillins among common CDI strains (32) could help explain the observed modestly elevated risk level for the penicillin class, such that the enhancement of CDI risk resulting from elimination of normal enteric flora is somewhat counterbalanced by anti-C. difficile activity. These discrepancies may also result from wide variations in the antibiotic spectrum of penicillin subclasses (including broad-spectrum penicillins used more in the hospital setting such as piperacillin-tazobactam). Our meta-analysis noted that tetracycline antibiotics have little antibiotic-associated risk, which is in keeping with the only meta-analysis of inpatient CDI (10) risk.
Like any observational study, the findings of studies incorporated into this meta-analysis could have been biased by methodological flaws, including issues of control selection, misclassification of both outcomes and exposures, and residual confounding (33). Our quality checklist attempted to assess the overall risk of these biases in each study; we outline some specific observations below. With respect to the definition of the population, two studies did not exclude patients exposed to hospital settings during the risk period, and as such may actually represent studies of community-onset but hospital-acquired disease (20, 27) while two studies were restricted to patients who received a C. difficile assay, and as such, the controls did not represent the source population of cases (20, 27). With respect to ascertaining the outcome, all positive cases may have been subject to misclassification due to infection with another diarrhea-causing organism. Further, in two studies (17, 20), a lack of clinical detail meant that hospital-diagnosed cases with onset of symptoms ≥48 h after admission could not be separated from those with onset within 48 h. As such, unmeasured inpatient antibiotic exposures may have caused the disease outcome. Indeed, in the study of Dial et al. (17), outpatient antibiotic exposures were detected in only 47.1% of the cases. Of the studies included in this meta-analysis study, only one (34) considered the robustness of results to diagnostic suspicion bias by comparing effect sizes from clinical diagnoses to those with test-based confirmation; they found no significant differences in effect with the clinically diagnosed subgroup.
Other potential sources of bias in our meta-analysis could include a lack of consensus regarding the appropriate time window for identification of antibiotic exposure. As risk associated with antibiotic exposure decreases with increasing time, larger effect sizes are liable to be found in studies looking at the shorter time windows. Indeed, the study in our primary analyses with the shortest exposure window reported larger effect sizes for all antibiotics except tetracyclines (17). In fact, the appropriate time window may differ between antibiotic classes due to differing antimicrobial effect duration (35). In addition, simultaneous administration of multiple antimicrobial agents and confounding by indication (as individuals receiving antimicrobials may have underlying health conditions placing them at greater risk for CDI) may have biased results.
Finally, although we did not find evidence to suggest that our findings were influenced by publication bias, we did notice some selective reporting of antibiotic class exposures. Specifically, the smallest study meeting the inclusion criteria (36) failed to report effect estimates for 3 of the 7 antibiotic classes (tetracyclines, macrolides, and sulfonamides and trimethoprim), and none of the studies reported on the impact of oxalizidenones, glycopeptides, carbapenems, or aminoglycosides.
In summary, and on the basis of the best available evidence, we found that the risk profiles for antimicrobial classes as risk factors for community-acquired CDI are similar to those described for health care-associated disease. In particular, antimicrobial classes with broad-spectrum, and potent anti-Gram-negative and/or antianaerobic bacterial activity, including cephalosporins, fluoroquinolones, and clindamycin, are most likely to cause CDI. In contrast, macrolides, penicillins, sulfonamides and trimethoprim, and particularly tetracyclines confer a lower risk of CDI. While community-acquired CDI remains fortunately less common than its health care-associated counterpart, we propose that CDI risk represents yet another factor that needs to be factored into the decision to prescribe antimicrobials (and the choice of antimicrobial) in the outpatient setting.
ACKNOWLEDGMENTS
We acknowledge Philippe Vanhems at the Édouard Herriot hospital in Lyon, France, for his guidance and considerate backing of this research partnership.
This work was supported by Kevin Brown's Frederick Banting and Charles Best Canada Graduate Scholarship from the Canadian Institutes of Health Research (CIHR). N. Khanafer received a grant for her dissertation research from Sanofi Pasteur, France. N. Daneman is supported by a Clinician Scientist Salary Award from CIHR. David Fisman received a grant from the Institute of Population and Public Health of the CIHR.
We declare that we have no conflicts of interest.
REFERENCES
1.
Kwong J, Crowcroft N, Campitelli M, Daneman N, Deeks S, and Manuel D. 2010. Ontario Burden of Infectious Disease Study (ONBOIDS): an OAHPP/ICES report. Ontario Agency for Health Protection and Promotion and Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada.
2.
Dubberke ER, Reske KA, Olsen MA, McMullen KM, Mayfield JL, McDonald LC, and Fraser VJ. 2007. Evaluation of Clostridium difficile-associated disease pressure as a risk factor for C. difficile-associated disease. Arch. Intern. Med. 167:1092–1097.
3.
Centers for Disease Control and Prevention. 2005. Severe Clostridium difficile-associated disease in populations previously at low risk–four states, 2005. Morb. Mortal. Wkly. Rep. 54:1201–1205.
4.
Bauer MP, Veenendaal D, Verhoef L, Bloembergen P, Van Dissel JT, and Kuijper EJ. 2009. Clinical and microbiological characteristics of community-onset Clostridium difficile infection in The Netherlands. Clin. Microbiol. Infect. 15:1087–1092.
5.
Rangaiah J, Wilks M, and Millar M. 2009. Community Clostridium difficile. BMJ 338:b1346. https://doi.org/10.1136/bmj.b2195.
6.
Al Saif N and Brazier J. 1996. The distribution of Clostridium difficile in the environment of South Wales. J. Med. Microbiol. 45:133–137.
7.
Rodriguez-Palacios A, Staempfli HR, Duffield T, and Weese JS. 2007. Clostridium difficile in retail ground meat, Canada. Emerg. Infect. Dis. 13:485–487.
8.
Miyajima F, Roberts P, Swale A, Price V, Jones M, Horan M, Beeching N, Brazier J, Parry C, Pendleton N, and Pirmohamed M. 2011. Characterisation and carriage ratio of Clostridium difficile strains isolated from a community-dwelling elderly population in the United Kingdom. PLoS One 6:e22804. https://doi.org/10.1371/journal.pone.0022804.
9.
Kuntz J, Chrischilles E, Pendergast J, Herwaldt L, and Polgreen P. 2011. Incidence of and risk factors for community-associated Clostridium difficile infection: a nested case-control study. BMC Infect. Dis. 11:194. https://doi.org/10.1186/1471-2334-11-194.
10.
Bignardi GE. 1998. Risk factors for Clostridium difficile infection. J. Hosp. Infect. 40:1–15.
11.
Thomas C, Stevenson M, and Riley TV. 2003. Antibiotics and hospital-acquired Clostridium difficile-associated diarrhoea: a systematic review. J. Antimicrob. Chemother. 51:1339–1350.
12.
Pai M, McCulloch M, Gorman JD, Pai N, Enanoria W, Kennedy G, Tharyan P, and Colford JM. 2004. Systematic reviews and meta-analyses: an illustrated, step-by-step guide. Natl. Med. J. India 17:86–95.
13.
Carratala J, Mykietiuk A, Fernandez-Sabe N, Suarez C, Dorca J, Verdaguer R, Manresa F, and Gudiol F. 2007. Health care-associated pneumonia requiring hospital admission: epidemiology, antibiotic therapy, and clinical outcomes. Arch. Intern. Med. 167:1393–1399.
14.
Sanderson S, Tatt ID, and Higgins JP. 2007. Tools for assessing quality and susceptibility to bias in observational studies in epidemiology: a systematic review and annotated bibliography. Int. J. Epidemiol. 36:666–676.
15.
Downs SH and Black N. 1998. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J. Epidemiol. Community Health 52:377–384.
16.
Vandenbroucke JP, Von Elm E, Altman DG, Gøtzsche PC, Mulrow CD, Pocock SJ, Poole C, Schlesselman JJ, and Egger M, STROBE Initiative. 2007. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med. 4:e297. https://doi.org/10.1371/journal.pmed.0040297.
17.
Dial S, Kezouh A, Dascal A, Barkun A, and Suissa S. 2008. Patterns of antibiotic use and risk of hospital admission because of Clostridium difficile infection. CMAJ 179:767–772.
18.
DerSimonian R and Laird N. 1986. Meta-analysis in clinical trials. Control Clin. Trials 7:177–188.
19.
Hirschhorn LR, Trnka Y, Onderdonk A, Lee ML, and Platt R. 1994. Epidemiology of community-acquired Clostridium difficile-associated diarrhea. J. Infect. Dis. 169:127–133.
20.
Levy DG, Stergachis A, McFarland LV, Van Vorst K, Graham DJ, Johnson ES, Park BJ, Shatin D, Clouse JC, and Elmer GW. 2000. Antibiotics and Clostridium difficile diarrhea in the ambulatory care setting. Clin. Ther. 22:91–102.
21.
Agresti A. 1999. On logit confidence intervals for the odds ratio with small samples. Biometrics 55:597–602.
22.
Davies HTO, Crombie IK, and Tavakoli M. 1998. When can odds ratios mislead? BMJ 316:989–991.
23.
Higgins JPT and Thompson SG. 2002. Quantifying heterogeneity in a meta-analysis. Stat. Med. 21:1539–1558.
24.
Smith G and Egger M. 2001. Going beyond the grand mean: subgroup analysis in meta-analysis of randomised trials, p 143–156. In Systematic review in health care: meta-analysis in context. BMJ Publishing Group, London, England.
25.
Viechtbauer W. 2010. Conducting meta-analyses in R with the metafor package. J. Stat. Softw. 36:1–48.
26.
Lowe DO, Mamdani MM, Kopp A, Low DE, and Juurlink DN. 2006. Proton pump inhibitors and hospitalization for Clostridium difficile-associated disease: a population-based study. Clin. Infect. Dis. 43:1272–1276.
27.
Wilcox MH, Mooney L, Bendall R, Settle CD, and Fawley WN. 2008. A case-control study of community-associated Clostridium difficile infection. J. Antimicrob. Chemother. 62:388–396.
28.
Stevens V, Dumyati G, Fine LS, Fisher SG, and Van Wijngaarden E. 2011. Cumulative antibiotic exposures over time and the risk of Clostridium difficile infection. Clin. Infect. Dis. 53:42–48.
29.
Aspevall O, Lundberg A, Burman LG, Åkerlund T, and Svenungsson B. 2006. Antimicrobial susceptibility pattern of Clostridium difficile and its relation to PCR ribotypes in a Swedish university hospital. Antimicrob. Agents Chemother. 50:1890–1892.
30.
Owens RC Jr, Donskey CJ, Gaynes RP, Loo VG, and Muto CA. 2008. Antimicrobial-associated risk factors for Clostridium difficile infection. Clin. Infect. Dis. 46:S19–S31.
31.
Dhalla IA, Mamdani MM, Simor AE, Kopp A, Rochon PA, and Juurlink DN. 2006. Are broad-spectrum fluoroquinolones more likely to cause Clostridium difficile-associated disease? Antimicrob. Agents Chemother. 50:3216–3219.
32.
Hecht DW, Galang MA, Sambol SP, Osmolski JR, Johnson S, and Gerding DN. 2007. In vitro activities of 15 antimicrobial agents against 110 toxigenic Clostridium difficile clinical isolates collected from 1983 to 2004. Antimicrob. Agents Chemother. 51:2716–2719.
33.
Egger M, Schneider M, and Smith GD. 1998. Meta-analysis spurious precision? Meta-analysis of observational studies. BMJ 316:140–144.
34.
Delaney JAC, Dial S, Barkun A, and Suissa S. 2007. Antimicrobial drugs and community-acquired Clostridium difficile-associated disease, UK. Emerg. Infect. Dis. 13:761–763.
35.
Pultz NJ and Donskey CJ. 2005. Effect of antibiotic treatment on growth of and toxin production by Clostridium difficile in the cecal contents of mice. Antimicrob. Agents Chemother. 49:3529–3532.
36.
Naggie S, Miller BA, Zuzak KB, Pence BW, Mayo AJ, Nicholson BP, Kutty PK, McDonald LC, and Woods CW. 2011. A case-control study of community-associated Clostridium difficile infection: no role for proton pump inhibitors. Am. J. Med. 124:276.e1-276.e7.
Information & Contributors
Information
Published In
Antimicrobial Agents and Chemotherapy
Volume 57 • Number 5 • May 2013
Pages: 2326 - 2332
PubMed: 23478961
Copyright
© 2013 American Society for Microbiology.
History
Received: 26 October 2012
Returned for modification: 27 November 2012
Accepted: 2 March 2013
Published online: 11 April 2013
Contributors
Metrics & Citations
Metrics
Note:
- For recently published articles, the TOTAL download count will appear as zero until a new month starts.
- There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
- Citation counts come from the Crossref Cited by service.
Citations
Citation
2013.
Meta-Analysis of Antibiotics and the Risk of Community-Associated Clostridium difficile Infection. Antimicrob Agents Chemother
57:.
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.